Week of 6 February 2023: Bitcoin and Ether fall back after a mostly rising month
When a rising market gives back part of its gains in a single week, what does the shape of that week say about whether the trend has changed or merely paused?
BTC fell from 22,932.91 to 21,783.54 (-5.01%) and ETH from 1,629.01 to 1,514.83 (-7.01%). Both highs came early on Wednesday 8 February and both lows late on Friday 10 February. The weekly return ranked at the 22nd percentile of the 260 weeks on record.
What the data shows. The week topped out quickly: BTC reached 23,452.00 at 01:00 UTC on Wednesday 8 February, and ETH its high of 1,697.91 in the same hour. From there both slid, with the lows arriving at 21:00 UTC on Friday 10 February (BTC 21,451.00, ETH 1,490.00). The worst day for both was Thursday 9 February (-5.08% for BTC, -6.37% for ETH), which also produced the largest hourly move: -2.14% for BTC and -2.59% for ETH in the hour starting 19:00 UTC. Trading activity was elevated, with BTC dollar volume 1.47 times its prior twelve-week average and ETH 1.16 times. Hourly correlation between the two was 0.92, and the ETH/BTC ratio fell 2.10%, so ETH fell somewhat harder within a broadly shared move. Both closes sat in the lower third of their weekly ranges (BTC 16.6%, ETH 11.9%).
What it does not show. Price data does not identify sellers or their reasons. Nothing here says whether the fall was profit-taking, forced liquidation, or a shift in expectations. The four prior weeks were mostly positive for BTC (21.86%, 8.79%, 4.56%, then -3.41%), so this week followed a strong run, but the data cannot establish whether the rally's driver has changed.
What a careful desk would check. Realised volatility, the size of actual swings scaled to a yearly figure, was 34.9% for BTC and 49.6% for ETH — moderate, and far from a stress week. The pattern to note is timing: highs early in the UTC day, lows late, and the largest hourly move in the same late window on Thursday. A desk would ask whether its risk limits assume the closing hours are quiet, and whether stops sized on the prior four mostly rising weeks were calibrated for a week where the market closed in its lower third.
Funding. BTC perpetual funding averaged 0.0089% per eight hours and never went negative (minimum 0.0023%); ETH averaged 0.0081%, though its minimum over any eight-hour window was -0.0025%. So longs paid shorts on average during the fall, but the rate stayed small (BTC maximum 0.01%), and for ETH the payment did not hold in every eight-hour window. The data shows the direction and size of the payment, not the size or composition of the positions behind it.
What would change this reading
- If a later week closes back above this week's BTC high of 23,452.00, the decline reads as a pause within the prior uptrend rather than a reversal.
- If BTC trades below Friday's low of 21,451.00, the lower-third close becomes the early signal of a deeper move.
- If ETH keeps falling faster than BTC, so that the ETH/BTC ratio drops beyond -2.10%, the move stops being broad and becomes an Ether-specific one.
- If funding stays positive while price falls, longs are paying to hold positions that are losing value, which is worth watching into the next week.
Terms
- realised volatility
- how much prices actually swung over a period, expressed as an annualised percentage
- perpetual funding
- a regular payment between long and short holders of a perpetual futures contract that keeps its price close to spot; positive means longs pay shorts
- percentile
- the share of past weeks with a lower value; the 22nd percentile means roughly four in five prior weeks were higher